USE OF ACTIVATED CHARCOAL FOR THE COLLECTION AND CONTAINMENT OF 133Xe

EXHALED DURING PULMONARY STUDIES

A. Liuzzi, J. Keaney, and G. Freedman Lowell Technological Institute, Lowell, Massachusetts, and Yale University School of Medicine, New Haven, Connecticut

The radioactive gas most widely used for pulmo (drierite) is connected integrally to the inlet to nary studies is 133Xe (./). Unless properly contained avoid moisture condensation on the charcoal. or vented, expired 133Xe will be a source of radia Vacuum pumping was used to draw gases through tion exposure for medical and paramedical personnel the trap system at a rate of approximately 6 liters/ working in the area of its use. Since 133Xeis soluble min which simulates the respiratory minute volume in fatty tissue (2) and consequently accumulates of normal humans. Tygon plastic tubing was used to with repeated exposures it can be an internal as well conduct gases to the trap and pump. Rubber tubing as an external source of radiation exposure. and fittings should be avoided because of the high In a previously reported method for collecting affinity of xenon for rubber (4). Cooling was 133Xe (3), liquid nitrogen is used to solidify xenon achieved by suspending the carbon trap in a poly on a suitable adsorbent screen. This method requires styrene bucket filled with a coolant. The coolants the availability of liquid nitrogen and an apparatus used were ordinary ice and water and dry ice with capable of withstanding moderate pressures. methanol. Figure 2 shows the trap with coolant The purpose of this paper is to present a simple bucket ready for use. method for the collection and containment of 133Xe Efficiency determinations. To determine the trap which takes advantage of the high adsorbability of ping efficiency for 133Xe from air with a flow rate activated charcoal for xenon. A trap containing acti vated charcoal as the adsorbent has been studied for Received Jan. 19, 1972; revision accepted Apr. 3, 1972. use in directly collecting 133Xe from the exhaled For reprints contact: Anthony Liuzzi, Dept. of Radio logical Sciences, Lowell Technological Institute, 1 Textile breath of patients. Efficiency of xenon collection and St., Lowell, Mass. 01854. containment were studied as a function of temper ature. 1.75"

MATERIALS AND METHODS Basic apparatus. The basic trap unit consists of a cylindrical glass vessel containing 100 gm of loosely packed activated carbon granules* with a glass tube running axially the length of the vessel and excluding the charcoal (Fig. 1). The central tube is connected to the inlet at the top of the vessel; the outlet is at the was then imaged with a Pho/Gamma III scintillation is conducted to the bottom of the vessel and passes up through the carbon to the outlet. Stopcocks are attached at the inlet and outlet. A moisture pretrap .25 ID X.28 OD Gloss Tubing consisting of a glass cylinder filled with CaSo4

* H-M-1 grade 6 GC 10/20 mesh, provided by courtesy FIO. I. Schematic of activated carbon trap with moisture pre of The Union Carbide Co. trap.

Volume 13, Number 9 673 LIUZZI, KEANEY, AND FREEDMAN

with a Nal(Tl) scintillation detector used in con junction with a count-rate meter and strip-chart recorder. The detector was collimated to view the first trap only. Air was pulled through the trap for periods of up to 2 hr after injection of 133Xe-laden saline into the mixing bulb and counting rates re corded by the strip-chart recorder. Times at which the activity in the trap was reduced by 10 and 50% of the maximum were determined from the chart re cordings for the various conditions of cooling. Redistribution of 133Xewithin the trap was studied by comparing scintiphotographs of the sealed-off trap taken after the initial 5-min collection period, under the various cooling conditions, and again 24 hr after the trap was removed from the coolant. In addi tion, the 24-hr redistributed trap was counted before and after a 5-min ventilation at the normal flow rate. FIG. 2. First trop with moisture pretrop arranged for cooling. The percentage of redistributed activity removed in 5 min was determined for each condition of col of 6 liters/min, a second identical carbon trap was lection. constructed. The two traps were then assumed to have the same xenon trapping efficiency at the same RESULTS temperature. Table 1 contains a summary of efficiencies and Efficiency measurements were performed by draw ing a 133Xe-air mixture through the two identical retention data for various conditions of trapping. Average efficiencies are given for 4, 6, and 2 deter traps in series for a suitable length of time and then minations, respectively, with the trap at room tem separately counting the activity in each trap under identical conditions. The 133Xe-air mixture was pro perature, cooled with ice and water and cooled with dry ice and methanol. In all cases, the efficiency duced by having a 3-in.-diam glass bulb whose exit exceeded 99%. Ventilation times for 10 and 50% was connected to the moisture trap of the first carbon trap by tygon tubing. Saline containing dissolved 133Xe removal were determined once for each trapping temperature. Retention is seen to increase at de was then injected with a syringe into the glass bulb creased coolant temperatures. through the entry opening while air was being drawn The percentage decrease in activity due to 5-min through the bulb and traps. Air flow was maintained ventilation after 24-hr room temperature redistribu- for 5 min after injection. Well within this time inter val a maximum counting rate is achieved in the first trap as monitored by a scintillation detector. After passing the 133Xe-air mixture through the two-trap TABLE 1. '":'Xe COLLECTION EFFICIENCIES AND system, stopcocks isolating each trap were closed RETENTION IN ACTIVATED CARBON and each trap removed from the system. Each trap Percent was then imaged with a Pho/Gamma III scintillation Ventilation times removed (min)* for 5-min |IIIMIJ camera to obtain integral counting rates as well as Collection Efficiency ventilationf scintiphotographs showing the spatial distribution of condition (%) 10% 50% (24-hr redist.) the activity within each trap. The traps were placed Room 99.5 ±00.2 13.0 17.0 0.59 at the center of the crystal field with the axis parallel temp. n = 4 to the face of the camera at a distance of 15 cm Ice and 99.9 ± 00.07 31.5 41.0 0.5o from the front of the collimator. water n = 6 The efficiencies of the traps (assumed to be equal) Dry ice 99.3, A = 00.4 63.0 90.0 0.57 were then calculated with the equation E = 1 — and n = 2 CZ/CH where E is the fraction of 133Xe removed methanol from the air passed through the traps, Ci the ob * Time after initial 5-min collection at which 10 or 50% of trapped activity is removed with the same flow rate (6 served net counting rate in the first trap, and C2 the liters/min). observed net counting rate in the second trap. t Percent of trapped activity removed during a 5-min flow Retention and redistribution of i:;:;\i- within the at 6 liters/min after trap was allowed to stand at room tem perature for 24 hr. trap. Retention of 133Xe within the trap was studied

674 JOURNAL OF NUCLEAR MEDICINE CONTAINMENT OF EXPIRED 133Xc

Direction of flow was taken after the trap was allowed to remain at room temperature for 24 hr; uniform redistribution is clearly evidenced. The lower scintiphotograph was taken subsequent to a 5-min ventilation at room temperature from the 24-hr redistributed trap rep resented in the middle scintiphotograph. One sees the results of migration of the xenon toward the exit side of the charcoal where xenon is leaving the trap.

DISCUSSION AND CONCLUSIONS Adsorption of xenon to activated charcoal is due mainly to Van der Waal's forces. Individual bound states of xenon and carbon will have finite lifetimes so that one would expect redistribution toward equi librium as seen experimentally in the redistribution experiment. With continued bulk flow through the trap, xenon will migrate in the direction of the bulk gas flow. This is clearly evident in Fig. 3. The effect of the activated carbon is to selectively slow down the net flow rate of the xenon. The important param eters, then, for effective collection at any given tem perature are the velocity of the bulk gas and the path length through the charcoal (for a given quality of activated charcoal). One would expect that there is a relationship between the migration velocity of the xenon and the velocity of the bulk gas, and the temperature of the gas and charcoal. These phe nomena account for the nonpermanent trapping as the xenon passes through the charcoal and for the fact that the xenon will eventually come off the char coal if gas continues to flow through the trap. Figure 4 shows a pilot setup for use of the trap system in a clinical perfusion study. A suitable (non- rubber) pulmonary mouthpiece is used and gases are conducted through a large-diameter tube to a pulmonary bag (upper right-hand corner in Fig. 2) and then to the trap system. The mouthpiece contains FIG. 3. Scintiphotogrophs showing distribution of Xc within two one-way check valves to direct flow in such a same charcoal trap (A) after 5-min collection with ice and water, way that the exhaled breath exits to the trap tube (B) after 24 hr at room temperature, and (C) after subsequent 5-min pull-off. Direction of flow was from left to right. only while the inhaled air enters from the room only. The pulmonary bag serves as a damping volume to accommodate the nonuniform flow of exhaled tion was also measured once for each trapping cool air. Nasal occlusion is necessary and may be ant condition. After redistribution has occurred, achieved as indicated in Fig. 4 or by the coopera retention is markedly reduced and is seen to be essen tion of the patient in exhaling through the mouth tially independent of the initial temperature condition only. The double-trap system has been used in a of trapping; approximately 57% is removed during number of perfusion studies with humans so that 5 min of flow at 6 liters/min. efficiencies could be determined and compared with Figure 3 shows scintigraphs of a trap at various the experimental results. The system was similarly times subsequent to the injection of xenon-laden used in a number of pulmonary experiments with saline into the experimental bulb. The upper scinti- dogs using a full face mask. In all cases the com photograph shows the distribution of xenon after a puted efficiencies were comparable to those obtained 5-min collection with the trap in ice and water; the experimentally. Indeed, in some patient applications activity is seen to be concentrated toward the inlet the efficiencies were somewhat better than those ob end of the charcoal. The middle scintiphotograph tained experimentally, most likely because collection

Volume 13, Number 9 675 LIUZZI, KEANEY, AND FREEDMAN

trapped and nonuniformly distributed will with time distribute throughout the trap so that in the first instance of flow the distal gas will begin to come off immediately. In the second instance long traps will effectively delay the time over which xenon is released to a vent system. The longer the trap, the longer the delay time. Although in a perfusion study nearly all of the injected xenon is removed in the exhaled breath within about 2 min, significantly long holdup times can be achieved with large traps. It is important to use a moisture pretrap since the exhaled breath is essentially 100% water-saturated. If this is not done, the charcoal can become moisture laden, thus decreasing the adsorbability of the carbon for xenon. An additional interesting feature of using activated charcoal to trap xenon from the exhaled breath is that it is quite possible to recover the xenon subse quently. Xenon can easily be removed from the activated carbon by heating. In a simple experiment, a xenon-laden trap was connected to a volumetric flask with all valves closed. The flask was evacuated, sealed off, and subsequently cooled with liquid nitro gen. The trap was heated by placing it in boiling water, and while the flask was in liquid nitrogen, the valve between the two components was opened. The valve was then closed after about 5 min and the FIG. 4. Breathing apparatus arrangement for patient use of trap counted. It was found that approximately 80% carbon trap system. Breathing bag, traps, and pump are located behind wall through which hose is led. of the xenon contained in the trap was transferred to the flask. This clearly indicates the feasibility of using or recirculating exhaled 133Xe. times for patients were generally shorter than those in the experimental situation. SUMMARY The results clearly indicate the feasibility of using From the results of this limited study of xenon activated charcoal as a means of controlling the trapping with activated charcoal it seems clear that escape of radioactive xenon used in clinical pulmo a relatively simple means of trapping, containing, nary studies. Since the trap works quite well at room and recovering exhaled radioactive xenon effluence temperature, a collection system may not require during pulmonary studies can be effected. the complexity associated with a cooling require ment. The charcoal trap can be used in essentially ACKNOWLEDGMENT two ways: for total trapping and containment, or as The laboratory work for this paper was done in the clini a delay line in conjunction with a venting system. cal laboratories of the Section of Nuclear Medicine, Dept. In the first case, individual nonradioactive traps of Radiology, Yale University, New Haven, Conn. would be used for each study with flow times limited REFERENCES to the order of 4 or 5 min for the trap configuration 7. LOKEN MK, WESTGATEHD: Using mXe and a scin used in these studies. Flow times could be extended tillation camera to evaluate pulmonary function. J NucíMed with longer traps and/or lower bulk flow velocities 9: 45-50,1968 related to the total mass of activated charcoal. Traps 2. Argon, Helium and the Rare Gases, vols I, II. Cook should not be reused until they have been cleared of G, ed, New York, Interscience, 1962 radioactive xenon either by storage and natural decay 3. CORRIGAN KE, MANTEL J, CORRIGAN HH: Trapping system for radioactive gas. Radiology 96: 571-573, 1970 or ventilation in a suitable facility. This follows since 4. KEANEY J, LIUZZI A, FREEDMAN OS: Large dose any xenon initially in the trap will migrate out with errors due to redistribution of mXe in carpules and plastic the bulk gas flow. In addition, as seen, xenon initially syringes. J NucíMed 12: 249-250, 1971

676 JOURNAL OF NUCLEAR MEDICINE